Classification of Apple Tree Disorders Using Convolutional Neural Networks

Nachtigall, Lucas Garcia; Araújo, Ricardo M.; Nachtigall, G. R.

doi:10.1109/ictai.2016.0078

Cited by 101 publications

(48 citation statements)

References 6 publications

(6 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Moreover, Nachtigall et al [18] made use of CNNs to detect and classify nutritional deficiencies and damage on apple trees. AlexNet was used as CNN architecture, so they made a comparison between Multilayer Perceptron (MLP) and the CNN, which was compared with seven volunteer experts.…”

Section: Related Workmentioning

confidence: 99%

Comparison of Convolutional Neural Network Architectures for Classification of Tomato Plant Diseases

et al. 2020

View full text Add to dashboard Cite

Tomato plants are highly affected by diverse diseases. A timely and accurate diagnosis plays an important role to prevent the quality of crops. Recently, deep learning (DL), specifically convolutional neural networks (CNNs), have achieved extraordinary results in many applications, including the classification of plant diseases. This work focused on fine-tuning based on the comparison of the state-of-the-art architectures: AlexNet, GoogleNet, Inception V3, Residual Network (ResNet) 18, and ResNet 50. An evaluation of the comparison was finally performed. The dataset used for the experiments is contained by nine different classes of tomato diseases and a healthy class from PlantVillage. The models were evaluated through a multiclass statistical analysis based on accuracy, precision, sensitivity, specificity, F-Score, area under the curve (AUC), and receiving operating characteristic (ROC) curve. The results present significant values obtained by the GoogleNet technique, with 99.72% of AUC and 99.12% of sensitivity. It is possible to conclude that this significantly success rate makes the GoogleNet model a useful tool for farmers in helping to identify and protect tomatoes from the diseases mentioned.

show abstract

Section: Related Workmentioning

confidence: 99%

Comparison of Convolutional Neural Network Architectures for Classification of Tomato Plant Diseases

et al. 2020

View full text Add to dashboard Cite

show abstract

“…The simulation result has shown that 90% of appropriate identification rate is achieved on apple disease leaf illustration and has shown that the proposed work model is efficient and feasible. Lucas G. Nachtigall et al [2] have studied the usage of CNN for automatically classifying and detecting the diseases, nutritional deficiencies and is being damaged by the apple trees herbicides from the leave images. This fundamental task is to ensure the enhanced quality of yields and is more executed by the scholars in the field that could lessen the scale and add the cost.…”

Section: Related Workmentioning

confidence: 99%

“…The general diseases are in the form of spots, which are yellow and brown. The other diseases are in the form of fungus, bacterial diseases, and virus [2]. Bin Liu et al identified the plant leaf disease by determining the illness of a plant.…”

Section: Introductionmentioning

confidence: 99%

Evolutionary Feature Optimization for Plant Leaf Disease Detection by Deep Neural Networks

Al-bayati¹,

Üstündağ²

2020

IJCIS

View full text Add to dashboard Cite

Apple leaf disease is the foremost factor that restricts apple yield and quality. Usually, much time is taken for disease detection with the existing diagnostic techniques; therefore, farmers frequently miss the best time for preventing and treating diseases. The detection of apple leaf diseases is a significant research problem, and its main aim is to discover an efficient technique for disease leaf image diagnosis. This article has made an effort to propose a method that can detect the disease of apple plant leaf using deep neural network (DNN). Plant diseases detection system (PDDS) architecture is designed. Speeded up robust feature (SURF) is used for feature extraction and Grasshopper Optimization Algorithm (GOA) for feature optimization, which helps to achieve better detection and classification accuracy. Classification parameters, such as Precision, Recall, F-measure, Error, and Accuracy is computed, and a comparative analysis has been performed to depict the effectiveness of the proposed work.

show abstract

“…Soft computing approaches have become the de facto approach for agricultural robotics as they have been found to be able to handle dynamic conditions reliably (Y. Huang et al, ). They have been utilised for a wide range of agricultural tasks: harvesting, yield estimation, weed‐spraying, pollination, and crop management (Bargoti & Underwood, a, b; Dias, Tabb, & Medeiros, ; Kurosaki et al, ; Nachtigall, Araujo, & Nachtigall, ; Sa et al, ; Wan, Toudeshki, Tan, & Ehsani, ; Wang, Song, & He, ; Zhang et al, ). Detection of apples (Bargoti & Underwood, b; Dias et al, ; Inthiyaz, Kishore, & Madhav, ; Moallem, Serajoddin, & Pourghassem, ; Prasad et al, ; Puttemans, Vanbrabant, Tits, & Goedemé, ; Soleimani Pour, Chegini, Zarafshan, & Massah, ) and strawberries (Habaragamuwa et al, ; Puttemans et al, ) has shown good results with detection rates up to 90% of the fruit under real‐world orchard conditions.…”

Section: Related Workmentioning

confidence: 99%

Improvements to and large‐scale evaluation of a robotic kiwifruit harvester

Williams

Ting

Nejati

et al. 2019

Journal of Field Robotics

View full text Add to dashboard Cite

The growing popularity of kiwifruit orchards in New Zealand is increasing the already high demand for seasonal labourers. A novel robotic kiwifruit harvester has been designed and built to help meet this demand [H. A. Williams et al. Biosystems Eng. 181 (2019), pp. 140–156]. Early evaluations of the platform have shown good results with the system capable of harvesting 51.0% of 1,456 kiwifruit in four bays with an average cycle‐time of 5.5 s/fruit. However, the harvester's high cycle‐time, and high fruit loss rate at 23.4%, prevent it from being commercially viable. This paper presents two new developments for the harvesting system, a new vision system and two new gripper variations designed to overcome the harvester's previous limitations. Furthermore, we have designed and conducted the largest real‐world evaluation of a robotic fruit harvesting system published to date with over 12,000 kiwifruit involved. The results of this trial demonstrated that our kiwifruit harvester is one of the most effective selective fruit harvesters in the world capable of successfully harvesting 86.0% of reachable fruit, and 55.8% of all kiwifruit with a cycle‐time of 2.78 s/fruit.

show abstract

Classification of Apple Tree Disorders Using Convolutional Neural Networks

Cited by 101 publications

References 6 publications

Comparison of Convolutional Neural Network Architectures for Classification of Tomato Plant Diseases

Comparison of Convolutional Neural Network Architectures for Classification of Tomato Plant Diseases

Evolutionary Feature Optimization for Plant Leaf Disease Detection by Deep Neural Networks

Improvements to and large‐scale evaluation of a robotic kiwifruit harvester

Contact Info

Product

Resources

About